Decision Fusion Methods in a Dispersed Decision System - A Comparison on Medical Data
A dispersed decision-making system that use knowledge accumulated in separate knowledge bases is considered in this paper. The system has complex and dynamic structure. In previous papers, different fusion methods were applied in this system. The novelty of this paper is to examine an approach in which many different fusion methods from the rank or from the measurement level are used simultaneously in one decision-making process. The obtained results were compared with the results obtained when fusion methods are used individually. In addition, the results were compared with an approach in which a dispersed system is not used, but the classifications generated based on each local base are directly aggregated using fusion methods. The conclusions indicated the legitimacy of the use of a dispersed system.
KeywordsDispersed system Fusion methods Combining classifiers
- 2.Gatnar, E.: Multiple-Model Approach to Classification and Regression. PWN, Warsaw (2008). (in Polish)Google Scholar
- 3.Ga̧golewski, M.: Data fusion: theory, methods, and applications. Institute of Computer Science Polish Academy of Sciences (2015)Google Scholar
- 9.Przybyła-Kasperek, M.: Comparison of four methods of combining classifiers on the basis of dispersed medical data. In: Czarnowski, I., Caballero, A.M., Howlett, R.J., Jain, L.C. (eds.) IDT 2016. SIST, vol. 57, pp. 3–13. Springer, Cham (2016). doi: 10.1007/978-3-319-39627-9_1CrossRefMATHGoogle Scholar
- 11.Przybyła-Kasperek, M., Nowak-Brzezińska, A.: Intersection method, union method, product rule and weighted average method in a dispersed decision-making system - a comparative study on medical data. In: Nguyen, N.-T., Manolopoulos, Y., Iliadis, L., Trawiński, B. (eds.) ICCCI 2016. LNCS, vol. 9876, pp. 451–461. Springer, Cham (2016). doi: 10.1007/978-3-319-45246-3_43CrossRefGoogle Scholar